Post by misty on Jul 17, 2007 10:03:37 GMT -5
It says: Please feel free to forward this newsletter to others you know who may be interested. So heres the full article:
In May 2000, the American Academy of Pediatrics issued a comprehensive set of guidelines for pediatricians to follow when evaluating children for ADHD. These guidelines emphasized that DSM-IV criteria should be carefully followed when making the diagnosis (see www.helpforadd.com/criteria.htm" for a review), and that information about the child's symptoms and functioning should be obtained directly from parent/caregivers and from teachers/school personnel to establish whether DSM-IV criteria are met. Although a variety of psychological tests are often used by clinicians in their ADHD evaluations, the AAP guidelines were explicit in stating that specific diagnostic tests should not be used routinely to establish the diagnosis of ADHD.
One type of test that is often used in ADHD evaluations is called a Continuous Performance Test (CPT). A CPT is a computer-based assessment in which the child is required to watch different stimuli flash on the screen, and to either respond or refrain from responding by pressing a particular key depending on the stimuli presented.
For example, in the Conners' Continuous Performance TEST (CCPT) the child is instructed to press the space bar for each letter that is briefly flashed except for the letter 'X'. The test lasts for 14 minutes and to do well the child must maintain their vigilant attention to a boring and repetitive task. If one starts to lose focus and day dream, it is easy to make errors of omission, i.e., not pressing the space bar when one is supposed to - or errors of commission, i.e., pressing it when one is not supposed to. Large numbers of the former are presumed to reflect problems with attention while high rates of the latter are suggested to reflect problems with impulsivity.
The scoring program provides a precise indication of the child's performance and how it compares to other children of the same age. As such, it is presumed to provide an accurate assessment of a child's ability to sustain their attention and to refrain from impulsive responding, and may thus add an 'objective' component to the evaluation of ADHD.
Many parents desire the use of such an objective measure when their child is being evaluated, and are concerned about a diagnosis being made on what may seem like an purely subjective assessments of their child's behavior. In my experience, this is particularly likely to be the case when parents have a different perspective on their child's behavior than their child's teacher, or when parents disagree with one another. In these instances, which are not uncommon, the availability of 'objective' measures like the Conners' CPT would seem to hold real value in helping to resolve such inconsistencies.
A study published in a recent issue of the Journal of Abnormal Child Psychology [Edwards, et al (2007). Estimates of the validity and utility of the Conners' Continuous Performance Test in the assessment of inattentive and/or hyperactive-impulsive behaviors in children. JACP, 35, 393-404] provides a thorough and comprehensive study of this important issue.
Participants were 104 6-12 year old children who were referred to an outpatient developmental center for the evaluation of attention and behavior problems. As part of the evaluation, each child's parent and teacher completed a standardized behavior rating scale to assess the child's symptoms of ADHD. Parents were also administered a structured psychiatric interview to assess for ADHD as well as for other psychiatric disorders. Children were administered a brief screening test of their intellectual ability and also completed the Conners' Continuous Performance Test.
The range of data obtained enabled the researchers to examine the association between children's performance on the Conners' CPT, parent report of ADHD symptoms, and teacher report of ADHD symptoms. Because parent and teacher reports are the foundation of a thorough evaluation for ADHD, the researchers were interested in whether the CCPT results showed significant associations with these reports.
In addition, the researchers determined whether each child was above a clinical cut-off for inattentive and hyperactive-impulsive symptoms using established national norms for parent ratings, teacher ratings, and CCPT results. This allowed them to examine how often these different methods of identifying children in a clinically elevated range for ADHD behaviors agreed with one another. As you will see below, the results are quite instructive.
- Results -
* Agreement Between Parent and Teacher Reports *
The authors first examined the agreement between parent and teacher report of ADHD symptoms. For inattentive symptoms, the correlation was statistically significant but relatively modest at r=.25 (the value of the correlation coefficient can range from -1 to 1 and a value of .25 would be considered to be of small magnitude.) For hyperactive impulsive symptoms, the correlation was higher at.49, which would generally be considered to reflect a moderately strong association.
The authors also looked at the percent of times that parent and teacher scores were on the same side of a clinical cut off and found it to occur in 65% of the cases for inattentive symptoms. In other words, parent and teacher ratings were either both above the clinical cut-off or both below the clinical cut-off for 65% of the children who were assessed. For hyperactive-impulsive symptoms the results were quite similar at 63% of the time.
This means, however, that for over 1/3 of the children, parent and teacher ratings were on opposite sides of what would be considered a clinically elevated range. This was true regardless of whether parent results were based on the standardized behavior rating scale score or on the structure psychiatric interview. These results highlight that agreement between parents and teachers in regards to a child's ADHD symptoms is often quite modest.
Of course, it is the frequent discrepancy between parent and teacher ratings that has motivated, in part, professionals' use of objective measure of attention and impulsivity such as the Conners' CPT in ADHD evaluations. The authors were thus quite interested in testing how strongly CCPT results related to parent and teacher reports.
* Agreement Between Conners' CPT Results and Parent/Teacher Reports *
The authors began by examining whether children's CCPT results were significantly correlated with parent and teacher ratings of ADHD symptoms. The results from the CCPT that were examined included errors of ommission (presumed to reflect attention problems), errors of commission (presumed to reflect problems with behavioral inhibition/impulsivity), as well as an overall index scores that combines multiple aspects of test performance.
After controlling for children's age and IQ score - which are both related to CCPT performance - there were no significant correlations between the CCPT Overall Index score and parent and teacher ratings of inattentive and hyperactive-impulsive behavior. There were also no significant correlations between the CCPT Omissions Score and parent and teacher ratings. The CCPT Commission score was found to be significant correlated with teacher ratings of inattentive and hyperactive-impulsive behaviors, but this was in the OPPOSITE direction of what one would predict, i.e., children with more errors of commission had lower teacher ratings of ADHD behaviors.
Next, the authors looked at how often a child scoring above or below the clinical cut-off on the different CCPT scales was found to score above or below the clinical cut-off according to the parent interview and the parent and teacher behavior ratings. Although the analyses conducted to examine this were extensive and complicated, the basic conclusion was that the level of agreement between CCPT results and parent and teacher results was not much better than chance. Even when results of the parent interview, parent behavior ratings, and teacher behavior ratings were all consistent with respect to whether or not the child had ADHD behaviors in a clinically elevated range, consistency with CCPT findings were barely better than chance.
Although the CCPT performed a bit better at matching classification based on parent and teacher reports than would a completely random test, the level of agreement was poor to slight. In general, agreement was a bit better for hyperactive-impulsive behaviors than for inattentive behaviors.
- Summary and Implications -
Results from this study highlight the challenges that clinicians face when evaluating a child who presents with attention and behavior problems for ADHD. They also highlight the confusion that many parents experience in the course of having their child evaluated.
In many cases, the level of agreement between parent and teacher reports of the child's behavior will be moderate at best, and this can make it difficult to determine whether DSM-IV diagnostic criteria for ADHD are met.
Clinically, when parents and teachers have different views of the child's behavior, parents can have a difficult time accepting the clinician's judgment concerning whether or not their child has ADHD. This can be a confusing and stressful situation for parents, who want to do everything they can to help their child succeed, but feel confused about how to reconcile what may be strikingly different views of their child's behavior.
In theory, objective tests like the CCPT could be useful in these situations as they provide a measure of attention problems and impulsive responding that is not influenced by either parents' or teachers' observations. As this study demonstrates, however, CCPT results will often bear little association to how either parents or teachers perceive the child. Thus, rather than clarifying what may be a confusing diagnostic picture, CCPT results may add to the confusion instead.
What are the implications of these results for the use of Continuous Performance Tests to assist in the diagnosis of ADHD? In considering this question it is important to emphasize that the CPT evaluated here - the Conners' Continuous Performance Test - is only one of several such tests that are widely used by clinicians. Other CPTs are set up differently and are thus likely to perform differently than the CCPT used in this study. One must be careful, therefore not to assume that the results obtained here would also be found for other CPTs.
Regarding the CCPT, however, results from this study highlight the need to be cautious when using this test as part of a comprehensive evaluation for ADHD. As discussed above, even in cases where parents and teachers are consistent with respect to whether they rate the child's ADHD symptoms as falling in a clinically elevated range, the odds that results from the CCPT will coincide with these reports is not much better than a 50/50 proposition.
Clearly, this could introduce confusion into the diagnostic process rather reducing it. For example, if parent and teacher reports converge on an ADHD diagnosis, what is one to make of a CCPT result that falls in the normal range? Because ADHD - like all psychiatric disorders - is diagnosed based on the presence of behavioral symptoms that cause significant impairment in a child's functioning - it would not be appropriate to disregard the perspective on those who know the child best simply because the CCPT result is in the normal range. Similarly, when parents and teachers do not report sufficient symptoms to warrant and ADHD diagnosis, assigning this diagnosis because the CCPT result is elevated would also contradict current practice guidelines.
What about cases where parent and teacher reports are inconsistent with respect to a child's ADHD symptoms? Presumably, it is these situations where incorporating objective data provided by Continuous Performance Tests could be of value by providing new evidence that is consistent with either the parents' or teacher's perspective.
Although this seems like a reasonable way to use such data, I am not aware of any research that has examined whether incorporating CCPT data into the diagnostic process in this way is actually helpful. And, in order for CCPT results to resolve such inconsistencies in a valid manner, how a child performs on this test would need to provide valid information about some aspect of the child's functioning in the 'real world' that may not be not readily picked up by parents and teachers.
Until these critical issues are resolved, the authors' conclusion that when evaluating a child for ADHD, clinicians should obtain "...information from informants who know the child well using reliable methods...and should be cautious in how CCPT results are integrated into these other standard evaluations methods" is a sound one, and it is important to be aware of the limitations of this particular test.
In May 2000, the American Academy of Pediatrics issued a comprehensive set of guidelines for pediatricians to follow when evaluating children for ADHD. These guidelines emphasized that DSM-IV criteria should be carefully followed when making the diagnosis (see www.helpforadd.com/criteria.htm" for a review), and that information about the child's symptoms and functioning should be obtained directly from parent/caregivers and from teachers/school personnel to establish whether DSM-IV criteria are met. Although a variety of psychological tests are often used by clinicians in their ADHD evaluations, the AAP guidelines were explicit in stating that specific diagnostic tests should not be used routinely to establish the diagnosis of ADHD.
One type of test that is often used in ADHD evaluations is called a Continuous Performance Test (CPT). A CPT is a computer-based assessment in which the child is required to watch different stimuli flash on the screen, and to either respond or refrain from responding by pressing a particular key depending on the stimuli presented.
For example, in the Conners' Continuous Performance TEST (CCPT) the child is instructed to press the space bar for each letter that is briefly flashed except for the letter 'X'. The test lasts for 14 minutes and to do well the child must maintain their vigilant attention to a boring and repetitive task. If one starts to lose focus and day dream, it is easy to make errors of omission, i.e., not pressing the space bar when one is supposed to - or errors of commission, i.e., pressing it when one is not supposed to. Large numbers of the former are presumed to reflect problems with attention while high rates of the latter are suggested to reflect problems with impulsivity.
The scoring program provides a precise indication of the child's performance and how it compares to other children of the same age. As such, it is presumed to provide an accurate assessment of a child's ability to sustain their attention and to refrain from impulsive responding, and may thus add an 'objective' component to the evaluation of ADHD.
Many parents desire the use of such an objective measure when their child is being evaluated, and are concerned about a diagnosis being made on what may seem like an purely subjective assessments of their child's behavior. In my experience, this is particularly likely to be the case when parents have a different perspective on their child's behavior than their child's teacher, or when parents disagree with one another. In these instances, which are not uncommon, the availability of 'objective' measures like the Conners' CPT would seem to hold real value in helping to resolve such inconsistencies.
A study published in a recent issue of the Journal of Abnormal Child Psychology [Edwards, et al (2007). Estimates of the validity and utility of the Conners' Continuous Performance Test in the assessment of inattentive and/or hyperactive-impulsive behaviors in children. JACP, 35, 393-404] provides a thorough and comprehensive study of this important issue.
Participants were 104 6-12 year old children who were referred to an outpatient developmental center for the evaluation of attention and behavior problems. As part of the evaluation, each child's parent and teacher completed a standardized behavior rating scale to assess the child's symptoms of ADHD. Parents were also administered a structured psychiatric interview to assess for ADHD as well as for other psychiatric disorders. Children were administered a brief screening test of their intellectual ability and also completed the Conners' Continuous Performance Test.
The range of data obtained enabled the researchers to examine the association between children's performance on the Conners' CPT, parent report of ADHD symptoms, and teacher report of ADHD symptoms. Because parent and teacher reports are the foundation of a thorough evaluation for ADHD, the researchers were interested in whether the CCPT results showed significant associations with these reports.
In addition, the researchers determined whether each child was above a clinical cut-off for inattentive and hyperactive-impulsive symptoms using established national norms for parent ratings, teacher ratings, and CCPT results. This allowed them to examine how often these different methods of identifying children in a clinically elevated range for ADHD behaviors agreed with one another. As you will see below, the results are quite instructive.
- Results -
* Agreement Between Parent and Teacher Reports *
The authors first examined the agreement between parent and teacher report of ADHD symptoms. For inattentive symptoms, the correlation was statistically significant but relatively modest at r=.25 (the value of the correlation coefficient can range from -1 to 1 and a value of .25 would be considered to be of small magnitude.) For hyperactive impulsive symptoms, the correlation was higher at.49, which would generally be considered to reflect a moderately strong association.
The authors also looked at the percent of times that parent and teacher scores were on the same side of a clinical cut off and found it to occur in 65% of the cases for inattentive symptoms. In other words, parent and teacher ratings were either both above the clinical cut-off or both below the clinical cut-off for 65% of the children who were assessed. For hyperactive-impulsive symptoms the results were quite similar at 63% of the time.
This means, however, that for over 1/3 of the children, parent and teacher ratings were on opposite sides of what would be considered a clinically elevated range. This was true regardless of whether parent results were based on the standardized behavior rating scale score or on the structure psychiatric interview. These results highlight that agreement between parents and teachers in regards to a child's ADHD symptoms is often quite modest.
Of course, it is the frequent discrepancy between parent and teacher ratings that has motivated, in part, professionals' use of objective measure of attention and impulsivity such as the Conners' CPT in ADHD evaluations. The authors were thus quite interested in testing how strongly CCPT results related to parent and teacher reports.
* Agreement Between Conners' CPT Results and Parent/Teacher Reports *
The authors began by examining whether children's CCPT results were significantly correlated with parent and teacher ratings of ADHD symptoms. The results from the CCPT that were examined included errors of ommission (presumed to reflect attention problems), errors of commission (presumed to reflect problems with behavioral inhibition/impulsivity), as well as an overall index scores that combines multiple aspects of test performance.
After controlling for children's age and IQ score - which are both related to CCPT performance - there were no significant correlations between the CCPT Overall Index score and parent and teacher ratings of inattentive and hyperactive-impulsive behavior. There were also no significant correlations between the CCPT Omissions Score and parent and teacher ratings. The CCPT Commission score was found to be significant correlated with teacher ratings of inattentive and hyperactive-impulsive behaviors, but this was in the OPPOSITE direction of what one would predict, i.e., children with more errors of commission had lower teacher ratings of ADHD behaviors.
Next, the authors looked at how often a child scoring above or below the clinical cut-off on the different CCPT scales was found to score above or below the clinical cut-off according to the parent interview and the parent and teacher behavior ratings. Although the analyses conducted to examine this were extensive and complicated, the basic conclusion was that the level of agreement between CCPT results and parent and teacher results was not much better than chance. Even when results of the parent interview, parent behavior ratings, and teacher behavior ratings were all consistent with respect to whether or not the child had ADHD behaviors in a clinically elevated range, consistency with CCPT findings were barely better than chance.
Although the CCPT performed a bit better at matching classification based on parent and teacher reports than would a completely random test, the level of agreement was poor to slight. In general, agreement was a bit better for hyperactive-impulsive behaviors than for inattentive behaviors.
- Summary and Implications -
Results from this study highlight the challenges that clinicians face when evaluating a child who presents with attention and behavior problems for ADHD. They also highlight the confusion that many parents experience in the course of having their child evaluated.
In many cases, the level of agreement between parent and teacher reports of the child's behavior will be moderate at best, and this can make it difficult to determine whether DSM-IV diagnostic criteria for ADHD are met.
Clinically, when parents and teachers have different views of the child's behavior, parents can have a difficult time accepting the clinician's judgment concerning whether or not their child has ADHD. This can be a confusing and stressful situation for parents, who want to do everything they can to help their child succeed, but feel confused about how to reconcile what may be strikingly different views of their child's behavior.
In theory, objective tests like the CCPT could be useful in these situations as they provide a measure of attention problems and impulsive responding that is not influenced by either parents' or teachers' observations. As this study demonstrates, however, CCPT results will often bear little association to how either parents or teachers perceive the child. Thus, rather than clarifying what may be a confusing diagnostic picture, CCPT results may add to the confusion instead.
What are the implications of these results for the use of Continuous Performance Tests to assist in the diagnosis of ADHD? In considering this question it is important to emphasize that the CPT evaluated here - the Conners' Continuous Performance Test - is only one of several such tests that are widely used by clinicians. Other CPTs are set up differently and are thus likely to perform differently than the CCPT used in this study. One must be careful, therefore not to assume that the results obtained here would also be found for other CPTs.
Regarding the CCPT, however, results from this study highlight the need to be cautious when using this test as part of a comprehensive evaluation for ADHD. As discussed above, even in cases where parents and teachers are consistent with respect to whether they rate the child's ADHD symptoms as falling in a clinically elevated range, the odds that results from the CCPT will coincide with these reports is not much better than a 50/50 proposition.
Clearly, this could introduce confusion into the diagnostic process rather reducing it. For example, if parent and teacher reports converge on an ADHD diagnosis, what is one to make of a CCPT result that falls in the normal range? Because ADHD - like all psychiatric disorders - is diagnosed based on the presence of behavioral symptoms that cause significant impairment in a child's functioning - it would not be appropriate to disregard the perspective on those who know the child best simply because the CCPT result is in the normal range. Similarly, when parents and teachers do not report sufficient symptoms to warrant and ADHD diagnosis, assigning this diagnosis because the CCPT result is elevated would also contradict current practice guidelines.
What about cases where parent and teacher reports are inconsistent with respect to a child's ADHD symptoms? Presumably, it is these situations where incorporating objective data provided by Continuous Performance Tests could be of value by providing new evidence that is consistent with either the parents' or teacher's perspective.
Although this seems like a reasonable way to use such data, I am not aware of any research that has examined whether incorporating CCPT data into the diagnostic process in this way is actually helpful. And, in order for CCPT results to resolve such inconsistencies in a valid manner, how a child performs on this test would need to provide valid information about some aspect of the child's functioning in the 'real world' that may not be not readily picked up by parents and teachers.
Until these critical issues are resolved, the authors' conclusion that when evaluating a child for ADHD, clinicians should obtain "...information from informants who know the child well using reliable methods...and should be cautious in how CCPT results are integrated into these other standard evaluations methods" is a sound one, and it is important to be aware of the limitations of this particular test.
I was about to conquer the world but then I got distracted by something shiny.
